VirtualWorld

VirtualWorld is an immersive simulation project designed to showcase AI-powered self-driving cars within a virtual environment. The project incorporates machine learning algorithms to enable cars to navigate complex routes, avoid obstacles, and make autonomous decisions in real time.

Features

• Self-Driving Simulation: Virtual cars navigate autonomously through traffic, avoid obstacles, and make smart decisions based on environmental factors.
• Machine Learning Integration: AI models are trained to control car movements such as speed, steering, and lane-switching.
• Interactive Environment: A visually dynamic world with various road scenarios, including intersections, turns, and traffic signals.
• Customizable Settings: Users can configure traffic density, car speed, and simulation parameters to test different scenarios.

Tech Stack

• Frontend: JavaScript (vanilla), HTML5, CSS3
• Machine Learning: TensorFlow.js (for implementing the neural network)
• Simulation: Custom-built physics engine for realistic car movements

How to Run

1. Clone this repository:

   git clone https://github.com/username/VirtualWorld.git

2. Open the project folder in Visual Studio Code.

3. Install the Live Server extension in VSCode:

   • Go to the Extensions tab in VSCode and search for "Live Server."
   • Install the extension by Ritwick Dey.

4. Once installed, open the index.html file.

5. Right-click on the index.html file and select Open with Live Server.

6. The project will automatically open in your browser at http://localhost:5500.

Neural Network Configuration

• The neural network is trained using a dataset of simulated driving scenarios.
• Input: Camera view (pixels) and sensor data from the virtual environment.
• Output: Steering angles, speed control, and lane-switching decisions.
• Training: You can retrain the model by modifying the trainModel.js script, using TensorFlow.js.

Future Enhancements

• Implement multi-agent simulations with multiple self-driving cars interacting in the same environment.
• Add more complex road layouts, weather conditions, and dynamic obstacles (e.g., pedestrians).
• Introduce VR support for a more immersive user experience.

Contributing

Feel free to contribute to the project by creating a pull request or submitting an issue. Check out the open issues for areas that need improvement.

License

This project is licensed under the MIT License – see the LICENSE file for details.

---

Author: Hanafe Mira